Aluminum oxide assisted stereoselective rearrangement of a

Oct 1, 1980 - ... of the Corey Ylide with α-Sulfenyl-, α-Sulfinyl-, or α-Sulfonylenones. Angela M. Bernard, Angelo Frongia, Pier P. Piras, Francesc...
0 downloads 0 Views 395KB Size
J. Org. Chem. 1980,45,4530-4532

4530

Scheme I

17

18 X =GHR. 0

In conclusion, we have shown that methylated thiophene-Zcarboxylic acids can be easily dilithiated with LDA (2 equiv). Treatment of the resulting dianions with carbon-containing electrophiles provides a variety of homologues of the starting acids. It should be noted that acids 1-4, used in this study, are either commercially available or are readily accessible.2 Therefore, direct homologation of the appropriate methylated carboxylic acid offers a convenient and efficient method for the synthesis of various substituted thiophene-2-carboxylic acids which, in turn, can be transformed into a variety of substituted thiophenes by virtue of the inherent versatility of the carboxy group. It is therefore anticipated that our procedure will lead to an even broader use of the thiophene nucleus as a template in organic synthesis. Acknowledgment. We are indebted to Drs. R. F. Hirschmann and E. J. Cragoe for their encouragement and to Drs. J. B. Bicking, M. G. Bock, and R. L. Smith for many helpful discussions throughout the course of this investigation. We also thank Dr. W. C. Randall and his staff for elemental analyses and Mr. A. Augenblick for GC analyses. Registry No. 1, 1918-79-2; 2, 14282-78-1; 3, 23806-24-8; 4, 65613-27-6; 5, 74965-72-3; 9, 74965-73-4; 10, 74965-74-5; 11, 7496575-6; 12, 74965-76-7; 13, 74965-77-8; 14, 74965-78-9; methyl 5hexylthiophene-2-carboxylate,74965-79-0; 5-(2-hydroxy-2-methylpropyl)thiophene-2-carboxylicacid, 74965-80-3;methyl 4-methyl-5pentylthiophene-2-carboxylate,74965-81-4; 3-methyl-5-deuteriothiophene-2-carboxylic acid, 74965-82-5; 3-deuteriomethylthiophene-2-carboxylic acid, 74965-83-6; 3-ethylthiophene-2carboxylic acid, 74965-84-7; 3-methyl-5-(2-hydroxy-2-methylpropyl)thiophene-2-carboxylic acid, 74965-85-8; pentyl bromide, 110-53-2; 2-(2-bromoethyl)-l,3-dioxolane, 18742-02-4; acetone, 6764-1; water-d2, 7789-20-0; methyl iodide, 74-88-4; butyl bromide, 109-65-9.

Supplementary Material Available: Experimental details for the preparation of 5 and NMFt data of the produds (3 pages). Ordering information is given on any current masthead page. Norman P.Gould, Ta-Jyh Lee* Department of Medicinal Chemistry Merck Sharp & Dohme Research Laboratories West Point, Pennsylvania 19486 Received July 8, 1980

Scheme I1 0

I

Summary: Contrary to the general observation that thermally and photochemically induced cyclopropyl ketone to dihydrofuran arrangements take place with partial loss of the stereochemical identity of the starting cyclopropane, a characteristic shared by the closely related vinylcyclopropanes and vinyloxiranes, a case of totally stereospecific, alumina-assisted cyclopropyl ketone to dihydrofuran rearrangement at room temperature was observed. Sir: The thermal and photochemical isomerization of cyclopropyl ketones and imines to dihydrofurans and di-

L

&

R=C,H,

2b R=CH,

hydropyrroles has been known since Clokel first described the irreversible rearrangement of phenylcyclopropanimine to 2-phenyl-4,5-dihydropyrroleat 170 "C. Ever since the extension of this discovery to cyclopropyl aldehydes and ketones was reported by Wilson,2only a limited number of reports have appeared in the l i t e r a t ~ r ein , ~contrast ~ to the extensive kinetic, stereochemical, and theoretical studies of the closely related vinylcyclopropane to cyclopentene7 and vinyloxirane to dihydrofurad rearrangements. These three processes share in common the fact that the stereochemical identity of the starting cyclopropane derivative is not reflected in the five-memberedring product (see Scheme presumably due either to the intervention of diradical species" free to rotate, to what has been termed by Doering as continuous diradical transition states," or to the combination of four concerted [1,3] sigmatropic ~ h i f t s although ,~ the issue remains at present under intense debate. In addition, the stereomutation of cyclopropanes12and cyclopropyl ketones6 that would tend to randomize even further the stereochemistry of rearranged products has been shown to take place under the same thermal and photochemical conditions. In the present communication ~~

Aluminum Oxide Assisted Stereoselective Rearrangement of a Cyclopropyl Ketone to 4,5-Dihydrofuran

COOR

OCH,

~~

~

(1) Cloke, J. B. J. Am. Chem. SOC.1929,51, 1174. (2) Wilson, C. L. J. Am. Chem. SOC.1947,69,3002. Armitage, D. M. A.; Wilson, C. L. Ibid. 1959,81, 2437. (3) Dauben, W. G.; Shaffer, G. W. J. Org. Chem. 1969, 34, 2301.

(4) Zimmerman, H. E.; Boettcher,R. J.; Braig, W. J. Am. Chem. SOC.

1973,95, 2155. (5) Doering, W. v. E.; Birladenau, L. Tetrahedron 1973, 29, 499. (6)McGreer. D. E.: McKinlev. J. W. Can. J. Chem. 1973. 51. 1487. (7) Andrewa; G. D.iBaldwin,y. E. J. Am. Chem. SOC.1976,M; 6705,

6706.

(8) Vogel, E.; Gunther, D. Angew. Chem., Int. Ed. Engl. 1967,5,385. Pomelet, J. C.; Manisse, N.; Chuche, J. C. R. Hebd. Seances Acad. Sci., Ser. C 1970,270,1894. Paladini, J. C.; Chuche, J. Tetrahedron Lett. 1971, 4383.

(9) Wilcott, M. R., III; Cargle, V. H. J , Am. Chem. SOC.1967,89,723. (10) Mazzocchi, P. H.; Tamburin, H. J. J. Am. Chem. SOC.1975, 97,

554. (11) Doering, W. v. E.; Sachdev, K. J. Am. Chem. SOC.1975,97,5512. (12) Bereon, J. A; Pedemn, L. D.; Carpenter, B. K. J. Am. Chem. Soc. 1976, 98, 122. Baldann, J. E.; Carter, C. G. Ibid. 1979, 101, 1325 and

references cited therein.

0022-3263/80/1945-4530$01.00/0 0 1980 American Chemical Society

J. Org. Chem., Vol. 45, No. 22, 1980 4531

Communications Scheme I11

6

$y

with neutral alumina activity I in a column, with chloroform as solvent, elution after 72 h furnished 0.910 g of homogeneous (high-performance LC) material whose infrared, UV, and proton NMR spectra resembled closely, not unexpectedly, those of compound 5. Particular attention was paid to the C4 methyl and C5 oxymethyne signals. The former at 6 1.28 (d, J = 7.1 Hz) compared more favorably with 6 1.15 (d, J = 7.1 Hz) of trans compound 1116 than with the value of 6 0.51 (d, J = 7.1 Hz)

...-

CH30

a R=C,H5 8 J

R=CH,

II

12

of cis compound 12,16 in which the cis-methyl group is appreciably shielded by the vicinal phenyl substituent. The additional 0.13-ppm deshielding observed for the C5 Al,O, methyl group of 9 is probably the consequence of the de"O CH30 shielding effect of the vicinal carboxylic ester at C3. On 9 the other hand, the signal of 6 4.88 (d, J = 7.0 Hz) of the we would like to report what is to our knowledge not only oxymethyne at C5 is closer to the value of 6 4.86 (d, J = the first example of Lewis acid assisted cyclopropyl ketone 8.1 Hz) assigned to the C5 proton of trans compound 11 to dihydrofuran rearrangement at room temperature but than to 6 5.47 (d, J = 9.4 Hz) of the same proton of cis also, more importantly, the first such transformation to compound 12, which does not experience the shielding occur with complete retention of cyclopropane stereoeffect of the trans-phenyl ring. These data suggest the chemistry. trans-C4-methyl-C5-p-methoxyphenyl structure for comWhen purification of a syn-anti mixture of cyclopropyl pound 9.1a Not a trace of cis epimer 10 could be detected @-ketoesters 3 and 413J4 (Scheme 11) in a 1.9:l ratio, by high-performance LC or NMR methods, indicating that prepared in 30% yield from the bis(copper(I1) hexathe cyclopropyl ketone to dihydrofuran rearrangement had fluoroacetoacetonate)-catalyzedthermolysis of ethyl 2taken place in a hitherto unrecorded stereospecific fashion. diazo-3-oxobutyrate (2a) and p-methoxystyrene, was atQuenching of the 7 and 8 to 9 process by elution after tempted by means of column chromatography through 24 h yielded a mixture of these three products in a neutral alumina activity I11 in chloroform, the collected 1.01.40:2.08ratio, which became 1.0:3.17:9.92after 48 h fractions contained a variable proportion of a third comof contact time. This result shows that transformation of ponent not present in the initial mixture. When the 7 into 9 occurs appreciably faster than that of 8 into 9. contact time with neutral alumina activity I was increased Although detailed kinetic studies, presently in progress in to 24 h, cyclopropanes 3 and 4 disappeared completely our laboratory, will be required to assess the mechanistic whereas the third component was recovered in nearly contribution of aluminum oxide to this rearrangement, the quantitative yield. This fraction proved to be homogeoccurrence of the well-known stereomutation of cycloneous by high-performance liquid chromatography (LC) propanes12would suggest that, alternatively, compounds and displayed signals in the proton NMR spectrum at 6 5 and 9 are formed only from syn isomers 3 and 7, re1.23 (t, 3 H, J = 7.0 Hz), 2.20 (t, 3 H, J = 1.8 HZ),'~,'~ spectively, whereas anti compounds 4 and 8 undergo solely 2.62-3.40 (m, 2 H), 3.78 (s, 3 H), 4.10 (q,2 H, J = 7.0 Hz), stereomutation to 3 and 7 at a slower rate, presumably via 5.45 (dd, 1 H, J1= 10.0 Hz, J2= 8.0 Hz), and 6.98 (2 d, selective bond disconnection of the C1-C2 cyclopropane 4 H) attributable only to dihydrofuran 5." The transbond, in analogy with Baldwin's conclusions.' Yet the formation of 3 and 4 into 5 , formally a [1,3]sigmatropic formation of diradical speciesa to account for this stereoshift not previously recorded at room temperature under mutation would not be possible in the present situation Lewis acid catalysis, generally requires more demanding since the stereochemical identity of cyclopropanes 7 and conditions.6 8 would be, at least partially, lost. On the other hand, the For evaluation of the stereochemical outcome of this intervention of aluminum oxide assisted continuous direarrangement, the stereochemically defined cyclopropanes radical intermediates," symmetry-allowed p r o c e s ~ e sor ,~~ 7a,b and 8a,b were subjected to scrutiny. These were symmetry-forbidden migration allowed under the control prepared in 17% yield from diazo ketones 2a,b and of subadjacent orbitalsmcannot be ruled out at the present trans-anethole (6)14 (Scheme 111). When a mixture of 7a stage. A discussion in greater detail of the mechanistic and 8a in a 1.251 ratio (0.932 g) was brought into contact implications based on formal kinetic data of further model compounds will be forthcoming.

.

\ i ~ ~ ~ c 0 0 c ~)5

0

(13) Elemental analyses were performed by Drs. H. Malissa and G. Reuter in Elbach, Germany. (14) Cyclopropanes3,4,7, and 8 were characterizedspectroscopically as the correspondingmethyl esters 3b, 4b, 7b, and 8b and by comparison with closely related cyclopropanes reported earlier.6 Their isolation as pure diasteroisomers was rendered impoeaible in view of their unavoidable tendency to rearrange to the dihydrofuran system. (15) This signal contains a characteristic transannular coupling between the Cz methyl protons and the C4 methylene, typical of dihydrofurans such as 5.16 (16) Scribe, M. P.;Delgphine, M. M. C. R. Hebd. Seances Acad. Sci. 1965,261,160. (17) Anal. Calcd for C1SH,B04:C, 68.67; H, 6.92; 0,24.41. C, 68.72; H, 6.95; 0,24.36.

Acknowledgment. This investigation has been generously supported by Grant No. S1-0726 of the Consejo Nacional de Investigaciones Cientificas y Tecnolbgicas CONICIT, of Venezuela. Registry No. 1, 637-69-4; 2a, 2009-97-4; 2b, 24762-04-7; 3a, 74947-69-6; 3b, 74947-70-9; 4a, 74964-94-6; 4b, 74947-71-0; 5a, (18) Anal. Calcd for C16Hm04:C, 69.53; H, 7.30; 0,23.17. Found'* C, 69.58; H, 7.22; 0,23.02. (19) Woodward, R. B.; Hoffmann, R. Angew. Chem. 1969,81, 797. (20) Berson, J. A.;Salem, L. J.Am. Chem. SOC.1972,94, 8917.

4532

J. Org. Chem. 1980,45,4532-4533

74947-72-1;6, 104-46-1;7a, 74947-73-2;7b, 74947-74-3;8a, 7498476-2;8b, 74984-77-3; 9,74947-75-4; A&, 74947-76-5;5b, 1344-28-1.

Scheme I

Miguel E. Alonso, * Angelina Morales Centro de Quimica Instituto Venezolano d e Investigaciones Cientificas IVIC, Apartado 1827, Caracas 1010 A, Venezuela Received August 15, 1980

Photochemistry of Dimethylthioketene Dimers

Summary: Thioketones and dithioesters, using 1,3cyclobutanedithione and 3-mercapto-2,2,4-trimethyldithio-3-pentenoic acid P-thiolactone as models are found to undergo a-cleavage to give the diradical and carbene as reactive intermediates. Sir: During the past several years, considerable research has been directed toward an understanding of the chemistry of electronically excited thioketones.' Analyses of the data from this research have revealed that thioketones are capable of undergoing three general types of reactions from both lower and higher excited states: (a) photoreduction, (b) cycloaddition to multiple bonds, and ( c ) intramolecular cyclization. These reactions are also primary photoreactions of ketones. But, thioketones have not yet been reported to undergo one of the primary photoreactions of ketones, namely, Norrish type I a-cleavage. In an attempt to explore the possibility of type I reactions in thioketones, we have studied the photochemistry of tetramethyl-1,3-cyclobutanedithione(1) and 3-mercapto2,2,4-trimethyldithio-3-pentenoic acid P-thiolactone (2). The preliminary results are presented in this communication. The electronic absorption spectrum of 1 shows three bands due to the thiocarbonyl group around 500 nm (n,r*, c 22), 298 (r,r*, 409), and 227 (n,a*, 21600). That of 2 shows three bands around 460 nm (n,r*, c 12), 344 (r,r*c=s, 4500), and 244 ( ~ , r * ~3800). +, Irradiation of 1 (450-W medium-pressure mercury lamp with Pyrex filter; - 5 X M) in nonhydroxylic solvents (cyclohexane, benzene, diethyl ether) gives two major products. These were identified as 3 and 4.293 Irradiation in methanol under similar conditions yields an additional product 5 (a 1:l adduct) along with 3 and 4.3 All of these products are believed to originate from the lowest n,r* triplet state of 1 since selective excitation of the SI band (A > 450 nm; isolated by the Corning glass filter CS-3.71) and triplet sensitization (fluorenone, 2-acetylnaphthalene, benzil, and biacetyl) gave all these products. Also, con(1) de Mayo, I?. Acc. Chem. Res. 1976,9,52. (2)In addition to 3 and 4, we have isolated two products believed to be some type of dimers of the intermediate carbenes b and e. Characterization of these two products is hampered by the inability to obtain pure samples of these compounds for spectral identification. (3)All three compounds 3-5 gave satisfactory elemental analyses. Their spectral data are shown below. 3: IR (neat) 2960,2920,2860,1440, cm-'; 'H NMR (CDC13) 6 1.33 (s,12H), 1350,1290,1250,1120,990,880 1.382(8, 12 H), 3.720 (8, 2 H); I3C NMR (CDClJ 6 24.054(q),28.569(q), 63.251 (s), 65.649 (d), 281.536 ( 8 ) ; mass spectrum (70eV), m / e 346 (M'). Anal. Calcd for C18H,S,: C, 55.48;H, 7.57. Found: C, 56.08;H, 7.87. 4: IR (neat) 2970,2920,2860,1440,1360, 1220,1170,1060,960,900,840 cm-'; 'H NMR (CDCl,) 6 1.578 (s, 6 H),1.802 (s, 6 H); I3C NMR (CDCI,) 6 33.563 (q),33.723 (q),70.963 (s), 77.891 (s), 245.175 (s), 267.671 (8); mam spectrum (70eV), m / e 204 (loo%), 172 (17.9),141 (42.9),128 (59.5),113 (40.5),96 (%.I),86 (83.3),85 (48.8),81 (53.6),71 (57.1),59 (33.3),41 (35.7). Anal. Calcd for C8H12S3:C, 47.06;H, 5.92. Found: C, 46.71;H, 6.401. 5: 'H NMR (CDCI,) 6 1.352 (s, 3 H), 1.393 (8, 3 H)!1.593 (.s, 6 H),3.349 (a, 3 H),4.809 (s, 1 H). A careful search for 2 during the irradiation of 1 failed to show the presence of any 2.

-4

-e

Scheme I1

/

I I ihD &

-d SOLVENT

P

PRODUCT

Cyclohexane

1 (> 60'1.)

Methanol

-6 ( 80%)

sistent with this, the triplet quenchers alloocimene (47 kcal/mol) and cyclooctatetraene (